U.S. patent number 8,300,764 [Application Number 12/585,436] was granted by the patent office on 2012-10-30 for method and device for detecting placement error of an imaging plane of a radiographic image detector, as well as method and device for correcting images.
This patent grant is currently assigned to FUJIFILM Corporation. Invention is credited to Yoshitaka Yamaguchi.
United States Patent |
8,300,764 |
Yamaguchi |
October 30, 2012 |
Method and device for detecting placement error of an imaging plane
of a radiographic image detector, as well as method and device for
correcting images
Abstract
For a radiographic image detector which includes an imaging
plane including two-dimensional matrix of pixel sections, each
pixel section storing, when exposed to radiation, a charge
according to amount of the radiation, and which is used to be
exposed to radiation transmitted through the same subject each time
the detector is shifted along a predetermined axis of shift, an
inclination of the two-dimensional matrix relative to the axis of
shift of the radiographic image detector is detected. The
inclination is detected by applying radiation two times to the
detector at different radiation application positions effected by
the shift of the detector so that a common marker is imaged during
each radiation application; carrying out a reading operation after
each radiation application to acquire image data representing image
information of the marker; and detecting the inclination based on a
positional relationship between marker images represented by the
image data.
Inventors: |
Yamaguchi; Yoshitaka
(Kanagawa-ken, JP) |
Assignee: |
FUJIFILM Corporation (Tokyo,
JP)
|
Family
ID: |
42007259 |
Appl.
No.: |
12/585,436 |
Filed: |
September 15, 2009 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20100067773 A1 |
Mar 18, 2010 |
|
Foreign Application Priority Data
|
|
|
|
|
Sep 16, 2008 [JP] |
|
|
2008-236728 |
Aug 31, 2009 [JP] |
|
|
2009-200423 |
|
Current U.S.
Class: |
378/62; 378/98.8;
382/131 |
Current CPC
Class: |
G06T
7/73 (20170101); G06T 2207/30004 (20130101); G06T
2207/10116 (20130101); G06T 2207/30244 (20130101) |
Current International
Class: |
G01N
23/04 (20060101); G06K 9/38 (20060101); H05G
1/64 (20060101) |
Field of
Search: |
;378/28-32,37,51-56,62,91,98,98.6,98.8,98.12,162,163,189-192,204,205,207,210
;382/129,132,103,209,216-221,254,266,268,275,279,284,286,287,289,293-297 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
11-244270 |
|
Sep 1999 |
|
JP |
|
2006-156555 |
|
Jun 2006 |
|
JP |
|
Primary Examiner: Midkiff; Anastasia
Attorney, Agent or Firm: Edwards, Esq.; Jean C. Edwards
Neils PLLC
Claims
What is claimed is:
1. An image correction method using an inclination detected with a
placement error detection method to be used with a radiographic
image detector to detect placement error of an imaging plane of the
radiographic image detector, the imaging plane including a
two-dimensional matrix of pixel sections, each pixel section
storing, when exposed to radiation, an electric charge according to
an amount of the radiation, the radiographic image detector
outputting image data representing radiographic image information
of a subject acquired through a reading operation, the radiographic
image detector being used to be exposed to radiation transmitted
through the same subject each time the radiographic image detector
is shifted to a different position along a predetermined axis of
shift, the method detecting inclination of the matrix relative to
the axis of shift, the method comprising: applying radiation two
times to the radiographic image detector at different radiation
application positions effected by the shift of the radiographic
image detector so that a common marker is imaged during each
radiation application; carrying out the reading operation after
each radiation application to acquire image data representing
radiographic image information of the marker; and detecting the
inclination based on a positional relationship between marker
images represented by the image data acquired during each reading
operation, wherein the correction method comprises: applying
radiation transmitted through a subject to the radiographic image
detector more than one time with shifting the radiographic image
detector between the more than one time of radiation application;
carrying out the reading operation after each radiation application
to acquire image data representing radiographic image information
of the subject; and applying image processing to at least a part of
the image data acquired during each reading operation based on the
detected inclination to eliminate misalignment along a joint line
in an image of the subject due to the inclination, the misalignment
being generated when the image of the subject is formed by
combining the image data, wherein imaging for detecting placement
errors and imaging for performing image correction are executed
individually.
2. The method as claimed in claim 1, wherein the inclination
comprises an inclination in a plane containing exposure axes of the
radiation applied during the two times of radiation
application.
3. The method as claimed in claim 1, wherein the inclination
comprises an inclination in the imaging plane.
4. The method as claimed in claim 1, wherein the inclination
remains unchanged when the radiographic image detector is
shifted.
5. The method as claimed in claim 1, wherein the inclination
changes along with the shift of the radiographic image
detector.
6. An image correction method using a displacement detected with a
placement error detection method to be used with a radiographic
image detector to detect placement error of an imaging plane of the
radiographic image detector, the imaging plane including a
two-dimensional matrix of pixel sections, each pixel section
storing, when exposed to radiation, an electric charge according to
an amount of the radiation, the radiographic image detector
outputting image data representing radiographic image information
of a subject acquired through a reading operation, the radiographic
image detector being used to be exposed to radiation transmitted
through the same subject each time the radiographic image detector
is shifted to a different position along a predetermined axis of
shift, the method detecting a displacement of the matrix from a
predetermined position for the matrix when the matrix is exposed to
the radiation, the method comprising: applying radiation two times
to the radiographic image detector at different radiation
application positions effected by the shift of the radiographic
image detector so that a common marker is imaged during each
radiation application; carrying out the reading operation after
each radiation application to acquire image data representing
radiographic image information of the marker; and detecting the
displacement based on a positional relationship between marker
images represented by the image data acquired during each reading
operation, wherein the correction method comprises: applying
radiation transmitted through a subject to the radiographic image
detector more than one time with shifting the radiographic image
detector between the more than one time of radiation application;
carrying out the reading operation after each radiation application
to acquire image data representing radiographic image information
of the subject; and applying image processing to at least a part of
the image data acquired during each reading operation based on the
detected displacement to eliminate misalignment along a joint line
in an image of the subject due to the displacement, the
misalignment being generated when the image of the subject is
formed by combining the image data, wherein imaging for detecting
placement errors and imaging for performing image correction are
executed individually.
7. The method as claimed in claim 6, wherein the displacement
comprises a displacement in a direction perpendicular to the axis
of shift.
8. The method as claimed in claim 6, wherein the displacement
comprises a displacement in a direction parallel to the axis of
shift.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
The present application claims priority from Japanese Patent
Application No. 2008-236728, filed Sep. 16, 2008, and from Japanese
Patent Application No. 2009-200423, filed Aug. 31, 2009, the
contents of all of which are herein incorporated by reference in
their entirety.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for detecting inclination
and/or displacement of an imaging plane of a radiographic image
detector during recording of radiographic image information on the
detector, and a device for carrying out the method.
The present invention further relates to a method for applying, to
image data representing a radiographic image of a subject acquired
using the radiographic image detector, correction to eliminate
misalignment along a joint line in a subject image which
misalignment is generated when the subject image is formed by
combining the image data, and a device for carrying out the
method.
2. Description of the Related Art
A panel-shaped radiographic image detector has conventionally been
used in practice, as described, for example, in Japanese Unexamined
Patent Publication No. 2006-156555 (hereinafter ref erred to as
patent document 1). The radiographic image detector includes an
imaging plane having a two-dimensional matrix of pixel sections,
and when radiation which carries image information is applied, each
pixel section stores an electric charge depending on the amount of
the radiation. Then, the radiographic image detector outputs image
data representing the image information through a reading
operation. In general, each pixel section includes: a charge
generation layer for generating an electric charge when exposed to
radiation; a voltage application electrode for applying a voltage
to the charge generation layer; a charge collection electrode for
collecting the electric charge generated at the charge generation
layer; and a switching element for reading out the electric charge
collected by the charge collection electrode, which are formed
using, for example, a TFT (thin film transistor) active matrix
array.
The above-described panel-shaped radiographic image detector
typically has a quadrangular shape, i.e., a rectangular or square
shape, and is widely used to record image information carried by
radiation which has been transmitted through the subject and
applied to the radiographic image detector. The radiographic image
detector having the quadrangular shape may sometimes be used to
record a long image representing a long portion, such as the entire
spine, of the subject, such as a human body. In this case, the
radiographic image detector is shifted along a predetermined axis
of shift so that each time the radiographic image detector receives
radiation transmitted through a different portion of the same
subject.
When the panel-shaped radiographic image detector is used in this
manner, the reading operation is carried out each time the
radiation is applied (each time a radiographic image is recorded)
to acquire image data representing a radiographic image during each
reading operation. Subsequently, the sets of image data are
combined to acquire image data representing the long portion of the
subject. The method to acquire image data representing a long
radiographic image in this manner is described, for example, in
Japanese Unexamined Patent Publication No. 11(1999)-244270
(hereinafter referred to as patent document 2), in which a cassette
containing a phosphorescent screen is used as an example.
When the radiographic images are combined as described above,
misalignment along the joint line may be observed in the combined
image due to inclination of the imaging plane of the panel-shaped
radiographic image detector. There are several types of
"inclination of the imaging plane" causing this problem. Now, the
types of inclination are described in detail with reference to
FIGS. 10 and 11.
First, FIG. 10 schematically shows, at "a", a side view of a system
for recording (imaging) a radiographic image, which includes a
radiation source 100, a stand 101 for guiding a quadrangular
panel-shaped radiographic image detector D when it is shifted, and
an imaging plane 102 in the radiographic image detector D. In this
example, a grid 103 is recorded as the subject for the convenience
of explanation of the problem. That is, radiation 104 emitted from
the radiation source 100 and transmitted through the grid 103 is
applied to the imaging plane 102 of the radiographic image detector
D.
In this case, the quadrangular panel-shaped radiographic image
detector D is orientated such that the panel surface and one side
of the panel is parallel to a direction in which the stand 101
extends (the direction of arrow H), and is to be shifted in the
direction of arrow H. That is, in this case, the direction of arrow
H is the axis of shift. Then, first and second radiographic imaging
operations are carried out by applying the radiation 104
transmitted through the grid 103 to the radiographic image detector
D, which is stationary before and after being shifted.
One problem here is that the imaging plane 102 (and thus the
two-dimensional matrix of pixel sections forming the imaging plane)
may be inclined by an angle .alpha. with respect to the surface of
the panel due to assembly error, or the like, of the radiographic
image detector D. In such a case, radiographic images of the grid
103 imaged through the first and second application of the
radiation are distorted, as shown at "b" and "c" in FIG. 10. That
is, when the first and second recorded images are joined at an area
in the vicinity of the lower edge of the first image and an area in
the vicinity of the upper edge of the second image, the transverse
length of the subject is different between theses areas, and
misalignment is generated along the joint line.
It should be noted that, in this case, with the radiographic image
detector D being set as described above, the inclination angle
.alpha. of the imaging plane 102 relative to the panel surface is
the inclination angle of the two-dimensional matrix relative to the
axis of shift H.
Next, the other problem is described with reference to FIG. 11.
FIG. 11 schematically shows, at "a", a front view of the system for
recording (imaging) a radiographic image, which includes the stand
101, the imaging plane 102, and the grid 103, as with FIG. 10.
Although the radiation source is not shown in this drawing, the
radiation source is disposed to apply radiation along an exposure
axis which is perpendicular to the plane of the drawing.
The quadrangular panel-shaped radiographic image detector D is
orientated in the same manner as shown in FIG. 10 and is to be
shifted in the direction of arrow H. Then, first and second
radiographic imaging operations are carried out by applying the
radiation 104 to the radiographic image detector D, which is
stationary before and after being shifted.
The other problem here is that the two-dimensional matrix of the
pixel sections may be inclined by an angle .gamma. with respect to
the axis of the shift, i.e., the one side of the panel, in a plane
parallel to the surface of the panel (i.e., a plane parallel to the
plane of the drawing) due to assembly error, or the like, of the
radiographic image detector D. It should be noted that only some of
the pixel sections G are illustrated in FIG. 11. In such a case,
radiographic images of the grid 103 imaged through the first and
second application of the radiation are distorted, as shown at "b"
and "c" in FIG. 11. That is, when the first and second recorded
images are joined at an area in the vicinity of the lower edge of
the first image and an area in the vicinity of the upper edge of
the second image, misalignment which looks like faulting is
generated along the joint line.
It should be noted that, also in this case, with the radiographic
image detector D being set as described above, the inclination
angle .gamma. of the two-dimensional matrix relative to one side of
the panel is the inclination angle of the two-dimensional matrix
relative to the axis of shift H.
For example, in a case where the radiographic image detector D has
a size of 40 cm.times.40 cm and a distance from the radiation
source to the imaging plane (SID) is 180 cm, the misalignment along
the joint line between the combined images is about 0.5 mm at an
end of the image when the inclination angle .alpha. is 0.31
degrees, and is about 0.5 mm at an end of the image when the
inclination angle .gamma. is 0.07 degrees, which is fairly
pronounced.
The above-described problems relate to cases where the quadrangular
panel-shaped radiographic image detector is shifted precisely along
the axis of shift which is parallel to the surface and one side of
the detector, and the two-dimensional matrix of pixel sections is
inclined within such a radiographic image detector. However, even
when the two-dimensional matrix of pixel sections is not inclined
within radiographic image detector, i.e., even when the matrix is
formed parallel to the surface and one side of the quadrangular
panel-shaped radiographic image detector, the similar problems
occur when the radiographic image detector itself is inclined
relative to the axis of shift of the detector. FIGS. 12 and 13
illustrate such cases where the two-dimensional matrix is inclined
relative to the axis of shift of the detector by the inclination
angles .alpha. and .gamma., respectively, since the radiographic
image detector itself is inclined.
Further, in the above-described cases, the inclination of the
matrix remains unchanged when the radiographic image detector is
shifted. However, if the radiographic image detector is gradually
inclined when it is shifted, the inclination of the matrix changes
along with the shift of the radiographic image detector, and the
similar problems occur. FIG. 14 schematically illustrates such a
situation, and schematically shows, at "a", a front view of a
system for recording (imaging) a radiographic image, which includes
the stand 101, the imaging plane 102, and the grid 103, as with the
system shown in FIG. 10 (this is the same for FIGS. 13 and 14).
Although the radiation source is not shown in this drawing, the
radiation source is disposed to apply radiation along an exposure
axis which is perpendicular to the plane of the drawing.
It should be noted that, in this case, the matrix is inclined and
is laterally displaced along with the shift of the radiographic
image detector. This phenomenon is due to such factors that a guide
mechanism for guiding the radiographic image detector being shifted
has low accuracy, or that there is a relatively large clearance
between a guide rod and a guide member that slides along the guide
rod, for example, forming the guide mechanism.
Radiographic images of the grid 103 imaged during the first and
second application of the radiation in this case are as shown at
"b" and "c" in FIG. 14. Also in this case, the misalignment which
looks like faulting is generated along the joint line when the
first and second recorded images are joined at the area in the
vicinity of the lower edge of the first image and the area in the
vicinity of the upper edge of the second image.
Further, as described above, the problem of misalignment along the
joint line between the images occurs not only due to the
inclination of the matrix, but also due to displacement of the
matrix from a predetermined position during application of the
radiation. Now, the displacement is described in detail.
FIG. 15 schematically shows a situation where the displacement
occurs. FIG. 15 schematically shows, at "a", a side view of a
system for recording (imaging) a radiographic image, where the
numeral "100" denotes the radiation source. When images to be
combined are taken, the radiographic image detector D is
essentially placed, during the first and second application of the
radiation, at each of predetermined positions which overlap with
each other to some extent. However, if a mechanism for shifting the
radiographic image detector D, for example, is aged, the
radiographic image detector D may be displaced from the
predetermined position in a direction which is parallel to the axis
of shift H during each application of radiation. FIG. 15 shows an
example in which the radiographic image detector D is displaced
downward by a length .DELTA.y from the predetermined position for
the second application of the radiation.
Radiographic images of the grid 103 imaged during the first and
second application of the radiation in this case are as shown at
"b" and "c" in FIG. 15. In this case, the images are combined with
assuming that the position y.sub.0 (see the drawing) on the image
taken during the first imaging operation corresponds to the upper
end of the image taken during the second imaging operation.
Actually, however, the upper end of the image taken during the
second imaging operation is displaced by the length .DELTA.y, and
thus misalignment is generated along the joint line.
Further, the above-described displacement may occur not only in the
direction parallel to the axis of shift H but also in a direction
perpendicular to the axis of shift H. FIG. 16 schematically shows a
situation where such displacement occurs. FIG. 16 schematically
shows, at "a", a front view of a system for recording (imaging) a
radiographic image. Although the radiation source is not shown in
this drawing, the radiation source is disposed to apply radiation
along an exposure axis which is perpendicular to the plane of the
drawing.
When images to be combined are taken, the radiographic image
detector D is essentially placed, during the first and second
application of the radiation, at each of predetermined positions
which are aligned with each other in the direction perpendicular to
the axis of shift H. If, however, a mechanism for shifting the
radiographic image detector D is aged or the stand 101 (more
particularly, a rail for guiding the radiographic image detector D
being shifted) is bent, as shown in the drawing, the radiographic
image detector D may be displaced from the predetermined position
in the direction perpendicular to the axis of shift H during the
application of radiation. In the example shown in FIG. 16, the
radiographic image detector D is displaced from the predetermined
position for the second application of radiation to the right by a
length of .DELTA.x.
Radiographic images of the grid 103 imaged during the first and
second application of the radiation in this case are as shown at
"b" and "c" in FIG. 16. In this case, the images are combined with
assuming that the images taken during the first and second imaging
operations are aligned with each other along the transverse
direction, i.e., in the direction perpendicular to the axis of
shift H. Actually, however, the image taken during the second
imaging operation is displaced by the length of .DELTA.x, and thus
misalignment is generated along the joint line.
The above-mentioned patent document 2 discloses a method for
correcting misalignment between two images when the two images are
combined (in this case, misalignment due to a difference of
distances between the subject and the imaging plane), in which a
grid contained in a cassette is imaged together with the subject,
and the correction is carried out based on the grid image contained
in each image. This method, however, cannot detect the inclination
and/or displacement of the two-dimensional matrix which may occur
when the above-described quadrangular panel-shaped radiographic
image detector is used, and thus cannot correct misalignment
between the images based on such an inclination and/or
displacement.
SUMMARY OF THE INVENTION
In view of the above-described circumstances, the present invention
is directed to providing an image correction method that can
eliminate, through a simple operation, misalignment along a joint
line between combined images due to inclination and/or displacement
from a predetermined position of the two-dimensional matrix of a
radiographic image detector, and a device for carrying out the
method.
The invention is further directed to providing a placement error
detection method for detecting inclination and/or displacement from
a predetermined position of the two-dimensional matrix of the
radiographic image detector, and a device for carrying out the
method, in order to accomplish the above-described image correction
method.
A first aspect of the placement error detection method for
detecting placement error of an imaging plane of a radiographic
image detector according to the invention is to detect the
above-described inclination of the two-dimensional matrix of pixel
sections of the imaging plane relative to the axis of shift of the
radiographic image detector. Specifically, the placement error
detection method is to be used with a radiographic image detector
including an imaging plane, the imaging plane including a
two-dimensional matrix of pixel sections, each pixel section
storing, when exposed to radiation, an electric charge according to
amount of the radiation, the radiographic image detector outputting
image data representing radiographic image information of a subject
acquired through a reading operation, the radiographic image
detector being used to be exposed to radiation transmitted through
the same subject each time the radiographic image detector is
shifted to a different position along a predetermined axis of
shift, the method detecting inclination of the matrix relative to
the axis of shift, the method including: applying radiation two
times to the radiographic image detector at different radiation
application positions effected by the shift of the radiographic
image detector so that a common marker is imaged during each
radiation application; carrying out the reading operation after
each radiation application to acquire image data representing
radiographic image information of the marker; and detecting the
inclination based on a positional relationship between marker
images represented by the image data acquired during each reading
operation.
The inclination of the matrix relative to the axis of shift to be
detected with the method may be an inclination in a plane
containing exposure axes of the radiation applied during the two
times of radiation application, or an inclination in the imaging
plane.
The inclination of the matrix may remain unchanged when the
radiographic image detector is shifted, or may change along with
the shift of the radiographic image detector (the latter includes a
case where the inclination does not occur depending on the shift
position).
A second aspect of the placement error detection method for
detecting placement error of an imaging plane of a radiographic
image detector according to the invention is to detect the
above-described displacement of the two-dimensional matrix of pixel
sections of the imaging plane from a predetermined position for the
matrix when the matrix is exposed to the radiation. Specifically,
the placement error detection method is to be used with a
radiographic image detector including an imaging plane, the imaging
plane including a two-dimensional matrix of pixel sections, each
pixel section storing, when exposed to radiation, an electric
charge according to amount of the radiation, the radiographic image
detector outputting image data representing radiographic image
information of a subject acquired through a reading operation, the
radiographic image detector being used to be exposed to radiation
transmitted through the same subject each time the radiographic
image detector is shifted to a different position along a
predetermined axis of shift, the method detecting a displacement of
the matrix from a predetermined position for the matrix when the
matrix is exposed to the radiation, the method including: applying
radiation two times to the radiographic image detector at different
radiation application positions effected by the shift of the
radiographic image detector so that a common marker is imaged
during each radiation application; carrying out the reading
operation after each radiation application to acquire image data
representing radiographic image information of the marker; and
detecting the displacement based on a positional relationship
between marker images represented by the image data acquired during
each reading operation.
The displacement of the matrix to be detected with this method may
be a displacement in a direction perpendicular to the axis of
shift, or a displacement in a direction parallel to the axis of
shift.
A first aspect of the placement error detection device for
detecting placement error of an imaging plane carries out the first
aspect of the placement error detection method. The device
includes: radiation application means for applying, to the
radiographic image detector, radiation transmitted through a common
marker; shifting means for shifting the radiographic image detector
in a direction of the axis of shift; image data acquiring means for
acquiring image data from the radiographic image detector each time
the radiographic image detector is shifted and the radiation is
applied to the radiographic image detector; and calculation means
for calculating the inclination based on a positional relationship
between marker images represented by the acquired image data.
A second aspect of the placement error detection device for
detecting placement error of an imaging plane carries out the
second aspect of the placement error detection method. The device
includes: radiation application means for applying, to the
radiographic image detector, radiation transmitted through a common
marker; shifting means for shifting the radiographic image detector
in a direction of the axis of shift; image data acquiring means for
acquiring image data from the radiographic image detector each time
the radiographic image detector is shifted and the radiation is
applied to the radiographic image detector; and calculation means
for calculating the displacement based on a positional relationship
between marker images represented by the acquired image data. On
the other hand, a first aspect of the image correction method
according to the invention is an image correction method including:
after the inclination has been detected with the first aspect of
the placement error detection method, applying radiation
transmitted through a subject to the radiographic image detector
more than one times with shifting the radiographic image detector
between the more than one times of radiation application; carrying
out the reading operation after each radiation application to
acquire image data representing radiographic image information of
the subject; and applying image processing to at least a part of
the image data acquired during each reading operation based on the
detected inclination to eliminate misalignment along a joint line
in an image of the subject due to the inclination, the misalignment
being generated when the image of the subject is formed by
combining the image data.
A second aspect of the image correction method according to the
invention is an image correction method including: after the
displacement has been detected with the second aspect of the
placement error detection method, applying radiation transmitted
through a subject to the radiographic image detector more than one
times with shifting the radiographic image detector between the
more than one times of radiation application; carrying out the
reading operation after each radiation application to acquire image
data representing radiographic image information of the subject;
and applying image processing to at least a part of the image data
acquired during each reading operation based on the detected
displacement to eliminate misalignment along a joint line in an
image of the subject due to the displacement, the misalignment
being generated when the image of the subject is formed by
combining the image data.
A first aspect of the image correction device according to the
invention carries out the first aspect of the image correction
method. The device includes: image correction means for applying
image processing to at least a part of the image data acquired
during each reading operation based on the detected inclination to
eliminate misalignment along a joint line in an image of the
subject due to the inclination, the misalignment being generated
when the image of the subject is formed by combining the image
data.
A second aspect of the image correction device according to the
invention carries out the second aspect of the image correction
method. The device includes: image correction means for applying
image processing to at least a part of the image data acquired
during each reading operation based on the detected displacement to
eliminate misalignment along a joint line in an image of the
subject due to the displacement, the misalignment being generated
when the image of the subject is formed by combining the image
data.
In the first aspect of the placement error detection method for
detecting placement error of an imaging plane according to the
invention, radiation is applied two times to, for example, a
quadrangular panel-shaped radiographic image detector at different
radiation application positions effected by the shift of the
radiographic image detector along the axis of shift so that a
common marker is imaged during each radiation application, the
reading operation is carried out after each radiation application
to acquire image data representing radiographic image information
of the marker, and the inclination of the matrix of pixel sections
of the imaging plane is detected based on a positional relationship
between marker images represented by the image data acquired during
each reading operation. Thus, the inclination can easily be
detected.
In the second aspect of the placement error detection method for
detecting placement error of an imaging plane according to the
invention, radiation is applied two times to a radiographic image
detector at different radiation application positions effected by
the shift of the radiographic image detector along the axis of
shift so that a common marker is imaged during each radiation
application, the reading operation is carried out after each
radiation application to acquire image data representing
radiographic image information of the marker, and the displacement
of the matrix of pixel sections of the imaging plane is detected
based on a positional relationship between marker images
represented by the image data acquired during each reading
operation. Thus, the displacement can easily be detected.
It should be noted that the placement error detection method for
detecting placement error of an imaging plane according to the
invention is applicable not only to elimination of misalignment
along a joint line between combined images, but also to
determination of an amount of modification when the inclination
and/or displacement of the imaging plane is manually and physically
modified, for example.
On the other hand, the first aspect of the placement error
detection device according to the invention includes: radiation
application means for applying, to the radiographic image detector,
radiation transmitted through a common marker; shifting means for
shifting the radiographic image detector in a direction of the axis
of shift; image data acquiring means for acquiring image data from
the radiographic image detector each time the radiographic image
detector is shifted and the radiation is applied to the
radiographic image detector; and calculation means for calculating
the inclination based on a positional relationship between marker
images represented by the acquired image data. Thus, the device can
implement the first aspect of the placement error detection method
of the invention.
The second aspect of the placement error detection device according
to the invention includes: radiation application means for
applying, to the radiographic image detector, radiation transmitted
through a common marker; shifting means for shifting the
radiographic image detector in a direction of the axis of shift;
image data acquiring means for acquiring image data from the
radiographic image detector each time the radiographic image
detector is shifted and the radiation is applied to the
radiographic image detector; and calculation means for calculating
the displacement based on a positional relationship between marker
images represented by the acquired image data. Thus, the device can
implement the second aspect of the placement error detection method
of the invention.
The first aspect of the image correction method according to the
invention uses the above-described method, which allows easily
detecting the inclination of the matrix of pixel sections of the
imaging plane, to apply image processing to eliminate the
misalignment along the joint line in the combined image due to the
detected inclination. Thus, elimination of the misalignment along
the joint line can easily be achieved.
The second aspect of the image correction method according to the
invention uses the above-described method, which allows easily
detecting the displacement of the matrix of pixel sections of the
imaging plane, to apply image processing to eliminate the
misalignment along the joint line in the combined image due to the
detected displacement. Thus, elimination of the misalignment along
the joint line can easily be achieved.
The first aspect of the image correction device according to the
invention includes image correction means for applying image
processing to at least a part of the image data acquired during
each reading operation based on the detected inclination to
eliminate misalignment along a joint line in an image of the
subject due to the inclination, the misalignment being generated
when the image of the subject is formed by combining the image
data. Thus, the device can implement the first aspect of the image
correction method.
The second aspect of the image correction device according to the
invention includes image correction means for applying image
processing to at least a part of the image data acquired during
each reading operation based on the detected displacement to
eliminate misalignment along a joint line in an image of the
subject due to the displacement, the misalignment being generated
when the image of the subject is formed by combining the image
data. Thus, the device can implement the second aspect of the image
correction method.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a partial side view illustrating one example of a
radiographic image detector to be subjected to imaging plane
inclination detection;
FIG. 2 is a schematic diagram for explaining radiographic imaging
carried out for detecting the inclination of the imaging plane;
FIG. 3 is a schematic diagram for explaining states of recorded
radiographic images when the imaging plane of the radiographic
image detector is inclined and when the imaging plane is not
inclined;
FIG. 4 is a schematic side view illustrating recording (imaging) of
the radiographic images;
FIG. 5 is a schematic diagram for explaining states of recorded
radiographic images when the imaging plane of the radiographic
image detector is inclined and when the imaging plane is not
inclined;
FIG. 6 is a schematic diagram for explaining one example of an
image correction method according to the invention;
FIG. 7 is a schematic diagram for explaining another example of the
image correction method according to the invention;
FIG. 8 is a schematic diagram for explaining still another example
of the image correction method according to the invention;
FIG. 9 is a diagram illustrating coordinate systems of the image
before and after image correction;
FIG. 10 is a diagram for explaining a prior-art problem;
FIG. 11 is a diagram for explaining another prior-art problem;
FIG. 12 is a diagram for explaining still another prior-art
problem;
FIG. 13 is a diagram for explaining yet another prior-art
problem;
FIG. 14 is a diagram for explaining yet still another prior-art
problem;
FIG. 15 is a diagram for explaining yet still another prior-art
problem;
FIG. 16 is a diagram for explaining yet still another prior-art
problem;
FIG. 17 is a schematic diagram for explaining one example of the
image correction method according to the invention;
FIG. 18 is a diagram illustrating the schematic configuration of a
placement error detection device for detecting placement error of
an imaging plane and an image correction device according to one
embodiment of the invention;
FIG. 19 is a graph showing one example of the relationship between
a position of the imaging plane and the placement error; and
FIG. 20 is a diagram for explaining a placement error detection
method and the image correction method according to another
embodiment of the invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Hereinafter, embodiments of the present invention will be described
in detail with reference to the drawings.
First, one example of a radiographic image detector, to which a
placement error detection method for detecting placement error of
an imaging plane according to the invention is applied, is
described. FIG. 1 is a partial sectional view showing an area
around pixel sections of such a radiographic image detector D. As
shown in the drawing, the radiographic image detector D includes an
active matrix substrate 10 including a number of pixel sections G,
and a charge generation layer (photoelectric conversion layer) 6
having electromagnetic conductivity and a voltage application
electrode (bias electrode: common electrode) 7 connected to a high
voltage power supply (not shown), which are formed in this order on
the active matrix substrate 10.
The pixel sections G include switching elements for reading out
electric charges collected by the charge collection electrodes, as
will be described later, and one pixel section G is formed per
switching element. The large number of pixel sections G are
arranged in a two-dimensional matrix, i.e., arranged in the
transverse direction in FIG. 1 and in the direction perpendicular
to the plane of FIG. 1. When the charge generation layer 6 is
exposed to an electromagnetic wave, such as X-ray, the charge
generation layer 6 generates an electric charge (electron-hole)
therein. In other words, the charge generation layer 6 has
electromagnetic conductivity, and converts image information
carried by the radiation into electric charge information. The
charge generation layer 6 is formed, for example, of a-Se
consisting primarily of selenium.
The active matrix substrate 10 includes a glass substrate 1S, a
gate electrode 2G, a gate insulating film 15, an upper storage
capacitor electrode 13, a semiconductor layer 9, a source electrode
8, a drain electrode 5, an interlayer insulating film 12 and a
charge collection electrode 11, where the gate electrode 2G, the
gate insulating film 15, the source electrode 8, the drain
electrode 5, the semiconductor layer 9, and the like, form a thin
film transistor (TFT) 4, which serves as a switching element
(hereinafter referred to as a TFT switch 4). In the TFT switch 4,
the source electrode 8 and the drain electrode 5 are connected to
data wiring (not shown), which is electrode wiring arranged in a
grid pattern, and to the upper storage capacitor electrode 13. The
semiconductor layer 9 serves to provide contact between the source
electrode 8, the drain electrode 5 and the gate electrode 2G.
The glass substrate 1S is a support substrate and is formed, for
example, of an alkali-free glass substrate. The gate insulating
film 15 is formed, for example, of SiN.sub.x or SiO.sub.x. The gate
insulating film 15 is formed to cover the gate electrode 2G and the
lower storage capacitor electrode 14. A portion of the gate
insulating film 15 above the gate electrode 2G serves as a gate
insulating film of the TFT switch 4, and a portion of the gate
insulating film 15 above the lower storage capacitor electrode 14
serves as a dielectric layer of a charge storage capacitor. That
is, the charge storage capacitor is formed at an area where the
upper storage capacitor electrode 13 overlaps with the lower
storage capacitor electrode 14, which is formed in the same layer
as the gate electrode 2G.
The gate electrode 2G and the lower storage capacitor electrode 14
are formed on the glass substrate 1S. The semiconductor layer 9 is
formed above the gate electrode 2G via the gate insulating film 15.
The source electrode 8 and the drain electrode 5 are formed on the
semiconductor layer 9. The gate insulating film 15 is formed above
the lower storage capacitor electrode 14, and the upper storage
capacitor electrode 13 is formed above the gate insulating film 15
above the lower storage capacitor electrode 14.
The interlayer insulating film 12 is formed, for example, of a
photosensitive acrylic resin and serves to electrically isolate the
TFT switches 4 from each other. A contact hole 16 is formed in the
interlayer insulating film 12, so that the charge collection
electrode 11 connects to the upper storage capacitor electrode 13,
which is a signal extracting electrode, via the contact hole
16.
The charge collection electrode 11 is formed, for example, of a
transparent amorphous conductive oxide film, and is formed above
the source electrode 8, the drain electrode 5 and the upper storage
capacitor electrode 13. The charge collection electrode 11 and the
charge generation layer 6 are electrically connected to each other,
so that the electric charge generated at the charge generation
layer 6 can be collected at the charge collection electrode 11. The
charge collection electrode 11 is electrically connected to the
drain electrode 5 and the charge storage capacitor of the TFT
switch 4 via the contact hole 16 formed in the interlayer
insulating film 12 to collect the electric charge generated at the
charge generation layer 6 and output the electric charge to the
outside via the TFT and the data wiring (not shown). The charge
generation layer 6 is formed immediately above the charge
collection electrode 11 to pass the electric charge to the charge
collection electrode 11.
A potential grading member 17 is disposed in the contact hole 16
and an area around the contact hole 16 such that the potential
grading member 17 fills up the contact hole 16. The area around the
contact hole 16 (the area denoted by "P" in FIG. 1) refers to an
area within about 5 .mu.m from the edge of the contact hole. The
potential grading member 17 may be provided with a curvature that
eases the edge of the contact hole 16.
The potential grading member 17 may be formed of an organic resin
having a low dielectric constant and a coefficient of thermal
expansion that is approximately the same as that of the material
forming the charge generation layer 6, and specific examples
thereof include photosensitive resins, such as novolac resin, epoxy
resin, acrylic resin, urethane resin, polyester resin, polyimide
resin, and polyolefin resin.
The high voltage power supply (not shown) is connected between the
bias electrode 7 and lower storage capacitor electrode 14. The high
voltage power supply applies a voltage between the bias electrode 7
and the lower storage capacitor electrode 14, thereby generating an
electric field between the bias electrode 7 and the charge
collection electrode 11 via the charge storage capacitor. The
charge generation layer 6 and the charge storage capacitor are
electrically connected in series. Thus, while a bias voltage is
applied to the bias electrode 7, the charge generation layer 6
exposed to radiation, such as X-ray, generates an electric charge
(electron-hole pairs) therein. The electrons generated in the
charge generation layer 6 move toward the positive electrode and
the holes generated in the charge generation layer 6 move toward
the negative electrode. As a result, the electric charge is stored
in the charge storage capacitor.
The radiographic image detector D as a whole includes the
two-dimensionally arranged charge collection electrodes 11, the
charge storage capacitors individually connected to the charge
collection electrodes 11, and the TFT switches 4 individually
connected to the charge storage capacitors. Thus, by once storing
two-dimensional electromagnetic information in the charge storage
capacitors and sequentially scanning the TFT switches 4,
two-dimensional electric charge information can be read out as
electric image data (an image signal).
As described above, in the radiographic image detector D, the
electric charge is stored in each pixel section G. Herein, the
plane in which the pixel sections G are arranged, i.e., the plane
parallel to the charge generation layer 6, is referred to as the
imaging plane.
Next, embodiments of the placement error detection method for
detecting placement error of the imaging plane and an image
correction method according to the invention are described. First,
a case is described, where, among various types of placement
errors, in particular, the previously-described inclination of the
two-dimensional matrix of the pixel sections of the imaging plane
relative to the axis of shift of the radiographic image detector is
detected, and the misalignment along the joint line between the
combined images due to the inclination is eliminated based on the
detected inclination.
First, a marker is imaged with an imaging system having the basic
structure as shown in FIGS. 10 and 11. At this time, in place of
the grid 103 shown in FIG. 10, a subject plane S as shown at "1" in
FIG. 2 is placed, and two markers M1 and M2, which are spaced from
each other by a predetermined distance in the horizontal direction,
are held on the subject plane S. The radiographic image detector D
held by the stand 101 is placed behind the subject plane S, i.e.,
on the side opposite from the radiation source 100.
As shown at "2" in FIG. 2, the quadrangular (rectangular or square)
panel-shaped radiographic image detector D is shifted from the
upper position to the lower position. The radiation transmitted
through the markers M1 and M2 is applied to the stationary
radiographic image detector D before and after being shifted,
thereby taking two radiographic images of the markers M1 and M2.
The imaging operations at this time may be achieved, for example,
by swiveling the radiation source or expanding the radiation
exposure field to cover an area corresponding to two panels.
During the two radiographic imaging operations, the radiographic
image detector D is in the position indicated by the diagonally
right up hatching shown at "2" in FIG. 2 for the first imaging
operation, and is in the position indicated by the diagonally left
up hatching shown at "2" in FIG. 2 for the second imaging
operation, so that the markers M1 and M2 are contained in both of
the images taken by the two imaging operations. During these
imaging operations, the radiographic image detector D and the
subject plane S are set such that the center positions in the width
direction of the radiographic image detector D and the subject
plane S are aligned with the center of the exposure field of the
radiation (the center in the transverse direction). The two markers
M1 and M2 are set in positions at an equal distance from the center
of the exposure field.
When the first radiographic imaging operation has been done, the
above-described reading operation is carried out prior to the
second imaging operation to acquire image data representing a
radiographic image of the markers M1 and M2. Also, when the second
radiographic imaging operation has been done, the reading operation
is carried out to acquire similar image data.
Next, how the inclination angle .alpha. shown in FIG. 10 is found
is described. FIG. 3 shows, at "1", radiographic images represented
by the image data acquired through the two reading operations. The
upper radiographic image in the drawing is obtained by the first
imaging and reading operation, and the lower radiographic image is
obtained by the second imaging and reading operation. Both the
radiographic images have the markers M1 and M2 recorded thereon. If
the imaging plane 102 is inclined by the angle .alpha. relative to
the panel surface, as shown in FIG. 10, positions of the markers M1
and M2 are misaligned between these radiographic images. The
inclination of the imaging plane 102 occurs due to assembly error,
or the like, when the structure such as one shown in FIG. 1 is
assembled and fixed in a quadrangular panel housing (the same
applies to inclination by the angle .gamma., which will be
described later). In contrast, if the imaging plane is not inclined
by the angle .alpha., radiographic images shown at "2" in FIG. 3
are obtained. Thus, the angle .alpha. can be found from the
relationship between the images shown at "1" and "2" in FIG. 3.
The angle .alpha. is an angle in a plane which contains exposure
axes of the radiation applied during the two radiographic imaging
operations (the two times of radiation application), i.e., a plane
parallel to the plane of FIG. 10. This is also the same in the
example of FIG. 12 shown above. Further, in this embodiment, the
panel-shaped radiographic image detector D is placed such that the
surface of the detector is parallel to the axis of shift of the
detector (the direction of arrow H), and thus the angle .alpha. is
an angle of the two-dimensional matrix of the pixel sections G
arranged on the imaging plane 102 relative to the axis of shift of
the detector.
Here, for the images shown at "1" in FIG. 3, the distance between
the markers M1 and M2 in the image acquired by the first operation
is denoted by w.sub.1 and the distance in the image acquired by the
second operation is denoted by w.sub.2. Further, as shown in FIG.
4, the distance from the center of the imaging plane to the markers
M1 and M2 in the direction in which the radiation is applied is
denoted by d, and the distance from the radiation source 100 to the
center of the imaging plane is denoted by SID. Furthermore, as
shown in FIGS. 3 and 4, twice the distance from the center of the
imaging plane to a midpoint between the markers M1 and M2 on the
imaging plane 102 is denoted by FOV.sub.h. The relationships
between the above distances are represented by equation 1 below,
and equation 2 is obtained therefrom:
.times..times..times..times..alpha..times..times..times..times..times..al-
pha..times..times..times. ##EQU00001##
In this manner, the inclination angle .alpha. of the imaging plane
102 can be found from the above-described distances based on
equation 2. To be exact, the difference between w.sub.1 and w.sub.2
fluctuates depending on the positional relationship between (the
heights of) the radiographic image detector D and the radiation
source 100, and thus the derived angle .alpha. also varies.
However, usually d is sufficiently small relative to SID, and the
fluctuation can be ignored to find the angle .alpha. through
approximation by equation 2. It should be noted that the angle
.alpha. in the example shown in FIG. 12 can be found in the same
manner as described above.
Next, how the inclination angle .gamma. shown in FIG. 11 is found
is described. The angle .gamma. is an angle of the two-dimensional
matrix relative to the axis of shift of the detector (the direction
of arrow H) within the plane of the imaging plane 102. This is the
same in the above-described example shown in FIG. 13.
FIG. 5 shows, at "1", radiographic images represented by the image
data acquired through the two reading operations. The upper
radiographic image in the drawing is obtained by the first imaging
and reading operation, and the lower radiographic image is obtained
by the second imaging and reading operation. Both the radiographic
images have the markers M1 and M2 recorded thereon. If the
two-dimensional matrix of the pixel sections G is inclined by the
angle .gamma. relative to one side of the panel, as shown in FIG.
11, positions of the markers M1 and M2 are misaligned between these
radiographic images. In contrast, if the imaging plane is not
inclined by the angle .gamma., radiographic images shown at "2" in
FIG. 5 are obtained. Thus, the angle .gamma. can be found from the
relationship between the images shown at "1" and "2" in FIG. 5.
Namely, assuming a square grid with one side thereof formed by a
line segment connecting the centers of the markers M1 and M2 in
each mage, as shown by dashed lines in FIG. 5, and shifting one of
the radiographic images in the longitudinal and transverse
directions relative to the other such that the center positions of
the grid in the images are aligned to each other, the angle .gamma.
can be found from amounts of shift in the two directions. It should
be noted that the angle .gamma. in the example shown in FIG. 13 can
be found in the same manner as described above.
Next, a process to correct for distortion of the radiographic
images due to the inclination based on the thus found angles
.alpha. and .gamma. is described. As one example, assuming a case
where the imaging plane 102 is inclined in the direction of the
angle .alpha., as shown by solid lines at "1" in FIG. 6, and the
subject is imaged through the first and second imaging operations
in the state shown at "2" in FIG. 6. In this case, if the
radiographic image of the subject record on and read out from the
imaging plane 102 can be corrected to provide a radiographic image
which is imaged in the state shown at "3" in FIG. 6, the distortion
of the images due to the inclination by the angle .alpha. can be
eliminated, thereby preventing misalignment at the joint line
between the two images when they are combined. It should be noted
that, in this embodiment, the images are also corrected to
eliminate the distortion of the images due to the inclination in
the direction of the angle .gamma. of the imaging plane 102.
It is assumed that the shift of the radiographic image detector D
has repeatability, and the radiographic image detector D is shifted
during imaging of the subject similarly to when the angles .alpha.
and .gamma. are found.
Next, how parameters to be used in the correction are found is
described. In this embodiment, a transformation matrix is found
from the angles .alpha. and .gamma.. More specifically, four or
more representative points are set, and the transformation matrix
is found based on correspondence between the representative points
before and after transformation. First, in order to correct for the
image distortion due to the inclination of the imaging plane in the
direction of the angle .alpha., parameters which allow the four
representative points shown by black circles shown at "1" in FIG. 7
(which are assumed here to form a square in a correctly recorded
state) to be corrected as the four representative points as shown
at "2" in FIG. 7. The lengths of the distorted lower and upper
sides of the uncorrected square are denoted by w.sub.1 and w.sub.2,
respectively, and the lengths of lower and upper sides of the
corrected square are denoted by w.sub.1' and w.sub.2',
respectively. Then, the relationships expressed by equations 3 and
4 below are established. The variables other than w.sub.1, w.sub.2,
w.sub.1' and w.sub.2' are the same as those described
previously.
'.times..times..times..times..times..alpha..times..times..times.'.times..-
times..times..times..times..alpha..times..times..times.
##EQU00002##
Then, the relationship expressed by equation 5 below is obtained
from equation 3, and the relationship expressed by equation 6 below
is obtained from equation 4.
.times.'.times..times..times..alpha..times..times..times.'.times..times.-
.times..alpha..times..times..times. ##EQU00003##
Then, the value in the parenthesis on the right-hand side of each
of equations 5 and 6 is calculated based on the found angle .alpha.
and the known SID and FOV.sub.h, and the calculated values are
respectively used as parameters to transform w.sub.1 into w.sub.1'
and w.sub.2 into w.sub.2'. Using these parameters, a (3.times.3)
matrix H.sub..alpha. for transforming the image data representing
the two-dimensional radiographic image shown at "1" in FIG. 7 into
that shown at "2" in FIG. 7 is found.
Further, FIG. 8 shows, at "1" and "2", a relationship between a
radiographic image before being rotated and the radiographic image
rotated by the angle .gamma.. In this case, a (3.times.3) matrix
H.sub..gamma. for transforming the image data representing the
two-dimensional radiographic image shown at "1" in FIG. 8 into that
shown at "2" in FIG. 8 is found. The matrix for transforming an
image according to a rotational relationship can be found using a
conventionally known method.
Since the inclination of the imaging plane 102 in the two
directions of angles .alpha. and .gamma. is summing of linear
phenomena, the two types of inclination can be combined by
multiplication of the matrixes as shown by equation 7 below:
H=H.sub..alpha.H.sub..gamma. or H=H.sub..gamma.H.sub..alpha.
Equation 7
Then, the matrix H obtained by the multiplication of the matrixes
shown by equation 7 is used as the transformation matrix. As the
image data, which representing the radiographic image of the
subject obtained through the first imaging and reading operation,
is transformed, the transformed image data is free of the
distortion due to the inclination of the imaging plane 102 in the
directions of the angles .alpha. and .gamma.. The same result is
obtained when the image data representing the radiographic image of
the subject obtained through the second imaging and reading
operation is transformed. Thus, the two transformed radiographic
images can be combined to provide a combined long radiographic
image without misalignment along the joint line.
As a specific example of an image transformation process using the
above-described transformation matrix, a two-dimensional projective
transformation is described. As shown at "1" and "2" in FIG. 9,
coordinate systems before and after the two-dimensional projective
transformation using the transformation matrix H are referred to as
an xy coordinate system and an x*y* coordinate system,
respectively. Generally, the two-dimensional projective
transformation is expressed, in a homogeneous coordinate system, by
equation 8 below:
.times..times..times..times..times..times..times..times..times..times..ti-
mes. ##EQU00004##
It should be noted that the homogeneous coordinate system handles a
n-dimensional problem as a (n+1)-dimensional problem to simplify
and generalize the calculation. The transformation matrix H has
nine components, however, has eight degrees of freedom. The
transformation matrix H can therefore be found when correspondence
of at least four points are obtained (that is, two equations with
respect to the xy coordinates are obtained for correspondence of
each point).
When the transformation matrix H has been obtained, the original
image data I can be corrected to provide a corrected image data I'
according to the equation below: I'=HI
In the above-described embodiment, although the image data is
corrected based on the inclination of the imaging plane 102 by the
angles .alpha. and .gamma. to eliminate the image distortion due to
the inclination, such correction may not be carried out and the
position of the imaging plane 102 may manually be modified to
eliminate the inclination of the imaging plane 102 by the angles
.alpha. and .gamma.. Further, the position of the imaging plane 102
may automatically be modified based on the inclination angles
.alpha. and .gamma. with an imaging plane position modifying means,
which is incorporated in the radiographic image detector D.
As the marker usable in the invention, the above-mentioned grid is
also applicable, besides the markers M1 and M2 representing two
points.
Next, another embodiment of the image correction method of the
invention for correcting for image distortion due to the
previously-described inclination angles .alpha. and .gamma. and the
displacements .DELTA.y and .DELTA.x respectively shown in FIGS. 15
and 16 is described. In the following description, the inclination
angle .alpha. may particularly be referred to as a "pitching
inclination angle", and the radiographic image detector D may
simply be referred to as a "panel" for clarity and convenience of
explanation. Assuming that a point on an image taken with a panel
which is inclined and/or displaced is expressed in an x-y
coordinate system, and a position on a final corrected image is
expressed in an x''-y'' coordinate system, the image correction
method of this embodiment calculates which point (x'', y'') the
image data at the point (x,y) should be related to.
First, how trapezoidal distortion of a taken image, as shown in
FIG. 10, due to the inclination of the panel by the pitching
inclination angle .alpha. is corrected, as shown in FIG. 17, is
described. This correction is hereinafter referred to as "trapezoid
correction". In this example, the imaging plane is not inclined
relatively to the panel, and thus the pitching inclination angle
.alpha. of the panel is equal to the inclination angle .alpha. of
the imaging plane, i.e., the two-dimensional matrix of the pixel
sections. In FIG. 17, the "normal imaging plane" refers to an
imaging plane which is free of the pitching inclination angle
.alpha.. Assuming that a point on the normal imaging plane is
expressed in an x'-y' coordinate system, the trapezoid correction
calculates which point (x',y') the image data at the point (x,y)
should be related to. Here,
.alpha.: the pitching inclination angle
(-90.degree.<.alpha.<+90.degree.),
x,y: coordinates on the inclined panel (actual image data),
x',y': coordinates on the image transformed through the trapezoid
correction,
d: SID (Source Image Distance, i.e., a distance between the panel
and the radiation source),
W: image width, and
H: image height.
From FIG. 17, an enlargement/reduction factor of the image in the
x-direction due to the presence of the pitching inclination angle
.alpha. is as follows: d:d-y sin .alpha.=x':x Equation 9
Moving the origin of the coordinate system from a point on the
inclined imaging plane, which a normal line crossing through the
center of the radiation source 100 hits, to a point on the normal
imaging plane, which a normal line crossing through the center of
the radiation source 100 hits, i.e., the point (0,0) shown in FIG.
17, the following equation is obtained from equation 9 above:
d:d-(y-H/2)sin .alpha.=x'-W/2:x-W/2
Thus, the relationship between the point (x,y) and the point
(x',y') is as follows:
'dd.times..times..times..alpha..times..times..times.'.times..times.
##EQU00005##
To be more precise, the image may also be enlarged or reduced in
the y-direction; however, it is assumed here that the enlargement
or reduction of the image in the y-direction is sufficiently small
and can be approximated as y'=y.
Next, how image distortion due to the inclination angle .gamma.
shown in FIGS. 11 and 13 and the displacements .DELTA.y and
.DELTA.x respectively shown in FIGS. 15 and 16 is corrected is
considered. In this example, the displacements .DELTA.y and
.DELTA.x are expressed as .DELTA.x=t.sub.x and .DELTA.y=t.sub.y.
The conditions here are as follows:
(x,y): coordinates on the original image,
(x',y'): coordinates on the image transformed through the trapezoid
correction,
(x'', y''): coordinates on the final corrected image,
t.sub.x: translation in the x-direction,
t.sub.y: translation in the y-direction, and
.gamma.: rotational angle in the x-y plane.
The relationship between the point (x,y) before the trapezoid
correction transformation and the point (x',y') after the trapezoid
correction transformation is as expressed by equation 10 above.
Expressing the point on the final corrected image in the x''-y''
coordinate system, as described above, the relationship between the
point (x'', y'') in this coordinate system and the point (x',y')
after the trapezoid correction transformation is as follows:
''''.times..times..times..gamma..times..times..gamma..times..times..gamma-
..times..times..gamma..times.''.times..times. ##EQU00006##
As described above, applying the two transformation operations
expressed by equations 10 and 11, the image data at the point (x,y)
is transformed into the image data at the point (x'',y''). Thus, by
applying the correction achieved through the above-described two
transformation operations to transform the original image data
associated with the point (x,y), which has been acquired through
the reading operation carried out on the radiographic image
detector D during each radiation application, into the image data
associated with the point (x'',y''), and combining the corrected
image data to form a single long image, generation of the
misalignment along the joint line in the image can be prevented
Next, a device for carrying out the above-described placement error
detection method and the image correction method are described.
FIG. 18 illustrates the schematic configuration of a radiographic
imaging device 150, which includes a placement error detection
device for detecting placement error of the imaging plane and an
image correction device according to one embodiment of the
invention. The radiographic imaging device 150 is adapted to
acquire a long radiographic image representing a large part of the
subject N by sequentially taking radiographic images of adjacent
areas N1, N2, . . . , in the subject N using the single radiation
source 100 and the single radiographic image detector D, and
combining the acquired images.
Specifically, the radiographic imaging device 150 includes: the
radiation source 100, which emits radiation 104 through an emission
window 111 toward an exposure range defined by a collimator 112;
the radiographic image detector D including the imaging plane
(radiation detection surface) 102, which detects the radiation 104
transmitted through the subject N and applied to the imaging plane;
a detector shifting unit 20, which shifts the radiographic image
detector D along the subject N; and a source positioning unit 25,
which positions the radiation source 100 to provide desired
position and orientation of the emission window 111. In FIG. 18,
"Cr" denotes the central axis of the radiation 104 with the
exposure range defined by the collimator 112.
The basic configuration of the radiographic image detector D is as
is described previously with reference to FIG. 1. The radiographic
image detector D detects the radiation 104 transmitted through the
subject N and converts the radiation into an electric signal to
output image data representing a radiographic image of the subject
N.
The detector shifting unit 20 includes: two supporting columns 21,
which stand in the vertical direction (the direction of arrow Y in
the drawing) from a floor surface 5F to hold the radiographic image
detector D therebetween; and a shift mechanism 22, which shifts the
radiographic image detector D in the vertical direction, i.e., in
the longitudinal direction. The shift mechanism 22 may be formed by
a mechanism that supports the radiographic image detector D with a
conventionally known linear slide mechanism, etc., and shifts the
radiographic image detector D using a drive source, such as a
motor.
When radiographic imaging is carried out for acquiring the
radiographic images to be combined, the subject N is positioned
along the direction in which the radiographic image detector D is
shifted. Namely, radiographic imaging is carried out with the
subject N standing on the floor surface.
The source positioning unit 25 holds and moves the radiation source
100 so that the radiation source 100 faces the imaging plane 102 of
the radiographic image detector D with the subject N standing
between the radiation source 100 and the imaging plane 102, i.e.,
the radiation source 100 is oriented substantially in the direction
of arrow Z in the drawing. The source positioning unit 25 includes:
a supporting column 26 extending in the vertical direction from a
ceiling 5E; a ceiling base 27, which moves the supporting column 26
along the ceiling 5E in the direction of arrow Z in the drawing;
and a rotating mount 28, which engages with the supporting column
26 to be movable in the direction of arrow Y in the drawing and
rotatable about an axis that is perpendicular to the plane of the
drawing. The radiation source 100 is mounted on the rotating mount
28. In this manner, the radiation source 100 is movable in the
vertical direction (the direction of arrow Y in the drawing) and in
the transverse direction (in the direction of arrow Z in the
drawing), and is rotatable about an axis which passes through the
substantial center of the radiation source 100 and is parallel to
the X-axis in the drawing. The source positioning unit 25 may be
formed by conventionally known mechanisms, such as a linear slide
mechanism and a rotary mechanism, and a drive source, such as a
motor.
The radiographic imaging device 150 further includes a long-image
imaging control unit 30, which controls operations of the detector
shifting unit 20 and the source positioning unit 25. The long-image
imaging control unit 30 controls operation of the detector shifting
unit 20 so that the radiographic image detector D is sequentially
shifted to positions Q1, Q2, . . . , for taking radiographic images
along the direction in which the subject N extends. At the same
time, the long-image imaging control unit 30 controls operation of
the source positioning unit 25 to position the radiation source 100
such that the application direction of the radiation 104 emitted
from the radiation source 100 is oriented toward the imaging plane
102 of the radiographic image detector D when the radiographic
image detector D is positioned at each of the above positions. As
the radiation source 100 is driven in this state, radiographic
images of the adjacent areas N1, N2, . . . , in the subject N are
sequentially taken to obtain image data representing each of image
portions, which are to be combined to represent the entire subject
N, during each imaging operation.
The radiographic imaging device 150 further includes an image
combining unit 35, which combines the image data acquired during
the radiographic imaging operations to provide a long radiographic
image representing the entire subject N. The thus combined long
radiographic image is displayed on an image display unit 60, which
is formed, for example, by a CRT display device.
The entire operation of the radiographic imaging device 150 is
controlled via a console 70. Therefore, information of the subject
N, imaging conditions for acquiring a long radiographic image,
etc., are inputted to the console 70, and these information are
fed, for example, to the long-image imaging control unit 30, an
imaging adjusting unit (not shown) for setting the radiation
application range defined by the collimator 112, etc. The imaging
adjusting unit adjusts the position of the radiation source 100,
condition of the collimator 112, the position of the radiographic
image detector D, etc., during each radiographic imaging operation
so that radiographic images of a predetermined size to be combined
are acquired through, for example, four radiographic imaging
operations. Then, operations for taking four radiographic images
are carried out according to instructions inputted via the console
70.
The size of the four radiographic images taken through the four
imaging operations may be determined by defining the radiation
application range with the collimator 112, as described above, or
by adjusting the length and width of each image portion by cutting
out a portion of each radiographic image acquired during each
imaging operation.
Next, a process of detecting placement error of the imaging plane
of the radiographic image detector D carried out in this device is
described. First, a case where the process is automatically carried
out by an automatic placement error detection device 80 is
described. The automatic placement error detection device 80
includes: a calibration image input unit 81, which obtains image
data from the radiographic image detector D; a marker detection
unit 82, which receives the output from the calibration image input
unit 81; and a placement error detection unit 83, which receives
the output from the marker detection unit 82. The output from the
placement error detection unit 83 is fed to a parameter calculation
unit 84.
When the placement error detection of the imaging plane is carried
out, radiographic imaging operations and reading operations for
reading the radiographic images taken through the radiographic
imaging operations for the placement error detection are carried
out separately from usual radiographic imaging operations of the
subject, according to an imaging menu inputted via the console 70,
for example. A series of operations from these operations to
operations for obtaining correction parameters are referred to as
"calibration", and radiographic images acquired during the
calibration are referred to as "calibration images". During this
calibration, the radiographic image detector D is sequentially
shifted to the positions Q1, Q2, . . . , and the radiation 104
transmitted through a marker, such as the markers M1 and M2
described above, is applied to the radiographic image detector D
staying at each position.
At this time, radiographic imaging is carried out such that the
image of the markers M1 and M2 is commonly captured within the
overlapping area on the radiographic image detector D when it is
positioned at each of the two adjacent positions Q1 and Q2. This is
the same for other two adjacent positions Q2 and Q3, and Q3 and Q4.
In order to carry out the radiographic imaging in this manner, the
markers may be arranged at an appropriate interval in the vertical
direction such that the image of the common markers is captured
within any overlapping area on the radiographic image detector D
positioned at each of the two positions, or the positions Q1, Q2, .
. . , may precisely be defined in advance and the markers may be
placed in overlapping positions corresponding to each of the
positions Q1, Q2, etc.
When an instruction to take the calibration images is made via the
imaging menu, an imaging range of each radiographic image
containing the marker, width of each overlapping area, and framing
of each image may automatically be set to predetermined values.
Further, the above-described markers may be provided on a screen,
and when the screen is set in a predetermined receptacle for
imaging of the calibration images, a screen detection signal may be
generated, and the signal may serve as a trigger for displaying
various menus for imaging of the calibration images on a display
section of the console 70, for example.
During the imaging of the calibration images with the radiographic
image detector D being positioned at each of the positions Q1, Q2,
etc., reading operation is carried out on the radiographic image
detector D for each imaging operation, and image data representing
the calibration image containing the marker is outputted from the
radiographic image detector D. The calibration image input unit 81
of the automatic placement error detection device 80 receives the
image data and feeds the image data to the marker detection unit
82. The marker detection unit 82 detects the position of the marker
based on image data which is sequentially sent from the
radiographic image detector D when it is positioned at each of the
two adjacent positions (for example, the positions Q1 and Q2) and
receives the applied radiation (hereinafter, the two images
represented by these image data are referred to as "upper and lower
images"), and inputs the information indicating the marker position
to the placement error detection unit 83. In order to determine the
position of the marker in each calibration image, a known
technique, such as template matching, may be used.
As the placement error detection unit 83 receives the information
indicating the marker position, the placement error detection unit
83 detects the placement error of the imaging plane of the
radiographic image detector D at the two adjacent positions, i.e.,
the above-described inclination angles .alpha. and .gamma. and the
displacements .DELTA.y and .DELTA.x, based on the received
information. The placement error detection unit 83 inputs the
information indicating the placement error to a parameter
calculation unit 84. As the parameter calculation unit 84 receives
this information, the parameter calculation unit 84 calculates
parameters used in the image transformation from the inclination
angles .alpha. and .gamma. and the displacements .DELTA.y and
.DELTA.x, and inputs the parameters to the image correction unit
34.
Basically, usual radiographic imaging operations, i.e.,
radiographic imaging operations to sequentially take radiographic
images of the adjacent areas N1, N2, etc., in the subject N, which
are to be combined to form the long image, are carried out after
the above-described calibration has been completed. However, the
calibration may be carried out, as necessary, in the course of
usual radiographic imaging operations which is carried out on a
daily basis. During the usual radiographic imaging operations, the
image data sequentially sent from radiographic image detector D
when it receives the applied radiation at each of the two adjacent
positions (for example, the positions Q1 and Q2) is fed to the
image combining unit 35, where the image data are combined to form
the combined image, as described above. Before the image data are
combined, the image data are subjected, at the image correction
unit 34, to correction based on the above-described parameters to
eliminate image distortion due to the placement error of the
imaging plane.
This correction is achieved by the previously-described
two-dimensional projective transformation. Therefore, the
parameters are specifically the values of the (3.times.3)
transformation matrix used for the two-dimensional projective
transformation. Using the thus corrected image data to form the
combined image, generation of the misalignment along the joint line
in the image can be prevented, as has been described in detail
previously.
Besides the parameters described as examples above, shear factors,
etc., may be applied as the parameters to achieve more accurate
elimination of the image distortion. Namely, it is known in the
two-dimensional projective transformation that shear transformation
may occur depending on ratios of coefficients a, b, c and d of the
(3.times.3) transformation matrix, and the coefficients a, b, c and
d are called the shear factors. When the two-dimensional projective
transformation is carried out using these shear factors to take the
shear transformation into account, more reliable elimination of the
image distortion due to the placement error of the imaging plane
can be achieved. More detailed description of the shear
transformation and the shear factors is found in Fujio Yamaguchi,
"Zukei-Shori-Kougaku (graphical processing engineering)", published
by The Nikkan Kogyo Shimbun, Ltd., 1981, pp. 73-75.
In the above-described embodiment, the radiographic imaging
operations are carried out such that the image of the common marker
is captured in all the overlapping areas on radiographic image
detector D, i.e., the overlapping area corresponding to the
positions Q1 and Q2, the overlapping area corresponding to the
positions Q2 and Q3, and the overlapping area corresponding to the
positions Q3 and Q4, and the marker position is detected each time.
However, the radiographic imaging operations may be carried out
such that the image of the common marker is captured in some of the
overlapping areas (for example, the overlapping area corresponding
to the positions Q1 and Q2 and the overlapping area corresponding
to the positions Q3 and Q4), and only these marker positions may be
detected. In this case, the marker position in the remaining
overlapping area (the overlapping area corresponding to the
positions Q2 and Q3 in this example) may be interpolated from the
actually detected marker positions.
Further, the inclination angles .alpha. and .gamma. and the
displacements .DELTA.y and .DELTA.x with respect to the overlapping
area for which the imaging of the marker has not been carried out
can be interpolated from the actually detected marker positions as
well as the inclination angles .alpha. and .gamma. and the
displacements .DELTA.y and .DELTA.x calculated based on the marker
positions. The interpolation may be achieved with any of known
methods, such as linear interpolation or spline interpolation.
In this case, the inclination angles .alpha. and .gamma. and the
displacements .DELTA.y and .DELTA.x may vary depending on the shift
position (panel position) of the radiographic image detector D due
to lack of accuracy of a linear slide mechanism forming the shift
mechanism 22, for example. FIG. 19 shows an example of
characteristics in which the inclination angle .alpha., for
example, changes depending on the panel position. Therefore, in a
case where the inclination angle .alpha., etc., is obtained through
interpolation, as described above, it is desirable to carry out the
interpolation with taking the characteristics as shown in FIG. 19
into account.
Further, the transformation parameters for correcting for the image
distortion may be calculated from the placement error of the
imaging plane, such as the above-described inclination angle
.alpha. associated with the panel position. The thus calculated
parameters may be associated with the panel positions and stored in
a storage means in advance, and when a panel position is detected,
the stored parameters associated with the panel position may be
read out from the storage means to be used for the transformation,
without calculating the transformation parameters for each time the
transformation operation is carried out.
In stead of carrying out the above-described interpolation, several
values of the placement error of the imaging plane, such as the
inclination angle .alpha., may be calculated and the values may be
averaged to use the average value as the placement error of the
imaging plane for all the positions of the radiographic image
detector D.
The above-described case is where the placement error of the
imaging plane is automatically detected by the automatic placement
error detection device 80. The device shown in FIG. 18 also
includes a user-responsive placement error detection device 95.
Now, how the placement error of the imaging plane is detected with
the user-responsive placement error detection device 95 is
described.
As shown in FIG. 18, the user-responsive placement error detection
device 95 includes, in addition to the console 90, an image input
unit 91, a corresponding point input unit 92, a placement error
detection unit 93 and a placement error saving unit 94, which are
individually connected to the console 90. The image input unit 91,
which is a means to obtain the image data outputted from the
radiographic image detector D, causes the console 90 to input the
calibration images CP, which are as described above. The images may
be inputted on-line, as with the case of the automatic placement
error detection device 80, or may be inputted off-line by recoding
the image data in any of various disks and reading out the image
data.
The console 90 causes two of the inputted calibration images CP to
be displayed on a pixel display section to allow the user to input
the positions of the marker, such as the markers M1 and M2 (see
FIG. 3, etc.), contained in the calibration images CP as
corresponding points using the corresponding point input unit 92,
which is formed by a mouse, or the like. It should be noted that,
as the two calibration images CP, those taken with the radiographic
image detector D at the two adjacent positions Q1 and Q2 are used,
similarly to the previously-described case.
The placement error detection unit 93 detects the inclination
angles .alpha. and .gamma. and the displacements .DELTA.y and
.DELTA.x based on the marker positions indicated by the inputted
corresponding points, in the same manner as the detection by
placement error detection unit 83 of the automatic placement error
detection device 80. The inclination angles .alpha. and .gamma. and
the displacements .DELTA.y and .DELTA.x, which are the placement
error of the imaging plane, are stored and saved in the placement
error saving unit 94. Thereafter, when an instruction to carry out
the transformation is inputted to the console 90 at an appropriate
time, the console 90 reads out the placement error EP stored and
saved in the placement error saving unit 94, and inputs the
placement error EP to the parameter calculation unit 84.
In the following operations, the parameter calculation unit 84
calculates the parameters and the image correction unit 34 applies
the transformation based on the parameters to the image data which
represents the radiographic images acquired in usual radiographic
imaging operations, in the same manner as the previously-described
case of the automatic placement error detection device 80. Using
the thus transformed image data to form the combined image,
generation of the misalignment along the joint line in the image
can be prevented.
It should be noted that, in stead of storing and saving the
placement error in the placement error saving unit 94, the
parameters calculated by the parameter calculation unit 84 may be
stored and saved in a storage means, and when the transformation is
carried out by the image correction unit 34, the parameters to be
used may be read out from the storage means.
Next, functions of the user-responsive placement error detection
device 95 are described in detail with reference to FIG. 20.
Functions (1)-(14) listed below are primary functions of the
user-responsive placement error detection device 95. FIG. 20 shows
display screens which may be displayed on an image display means of
the console 90, for example, to allow the user to use any of the
functions (1)-(14). How these functions are implemented is
described below.
(1) Function to Select Inputted Images to be Displayed
When the user clicks on areas 1A on the display screen with a
mouse, or the like, identification numbers of the inputted images
are displayed, and then the user clicks on a desired identification
number to select the inputted image.
(2) Function to Display Positional Information (Height) of the
Imaging Plane Associated with the Image
The information is displayed at areas 2A on the display screen.
(3) Function to Separately Display the Upper and Lower Images
The images are displayed at areas 3A on the display screen.
(4) Function to Display 50% Reduced Images of the Upper and Lower
Images with the Overlapping Area
This function is used in place of function (3) above. The upper and
lower images, which are transparent to each other, are displayed in
an overlapping state at an area 4A on the display screen.
(5) Function to Enlarge or Reduce the Displayed Image
The user clicks on areas 5A on the display screen with the mouse,
or the like, to select enlarged display or reduced display.
(6) Function to Input the Corresponding Points Between the Upper
and Lower Images with Mouse Click
When user indicates a corresponding point with a cursor, as shown
at an area 6A on the display screen, and clicks on the
corresponding point with the mouse, or the like, the corresponding
point is inputted.
(7) Function to Automatically Detect the Corresponding Points, or
Function to Automatically Detect More Detailed Corresponding Points
Around the Inputted Corresponding Points
As the user clicks on an area 7A on the display screen with the
mouse, or the like, the corresponding points are automatically
detected.
(8) Function to Calculate the Parameters from the Detected
Placement Error and to Display the Image Corrected Using the
Parameters
The corrected image is displayed at an area 8A on the display
screen.
(9) Function to Select Transparency Ratios at the Overlapping Area
in the Displayed Corrected Image
When the user clicks on an area 9A on the display screen with the
mouse, or the like, transparency ratios are displayed, and then the
user clicks on a desired transparency ratio to select the
transparency ratio. The ratios are displayed, for example, as
"upper image 50%", "lower image 50%", etc. When a transparency
ratio R of one of the images is selectable within the range from 0%
to 100%, the transparency ratio of the other image is (100%-R).
(10) Function to Display the Placement Error
The placement error is displayed at an area 10A on the display
screen.
(11) Function to Allow Manual Adjustment of the Placement Error and
to Equally Reflect the Adjustment on the Corrected Images, or
Function to Allow Fine Adjustment of the Individual Values
The values are adjusted by the user placing a cursor on an area 11A
on the display screen and operating (such as rotating) a mouse
dial, or the like.
(12) Function to Detect and Update the Placement Error
(Selectable)
As the user clicks on an area 12A on the display screen with the
mouse, or the like, the placement error is detected and
updated.
(13) Initialization Function to Return the Placement Error to
Default Values (Selectable)
As the user clicks on an area 13A on the display screen with the
mouse, or the like, values of the placement error are
initialized.
(14) Function to Save the Detected Placement Error (Selectable)
As the user clicks on an area 14A on the display screen with the
mouse, or the like, the detected placement error is saved.
With the user-responsive placement error detection device 95 having
the above-described functions, any of five procedures (A)-(D) below
may be used to detect the placement error of the imaging plane.
(A) Automatic Corresponding Point Detection
Select the inputted images via the function (1). Automatically
detect the corresponding points (through template matching across a
predetermined range, for example). Detect the placement error from
the corresponding points. (B) Semiautomatic Corresponding Point
Detection Select the inputted images via the function (1). Input
the corresponding points on the reduced images via the function
(6). Automatically detect more detailed corresponding points around
the corresponding points (through template matching across a
predetermined range, for example). Detect the placement error from
the corresponding points. (C) Manual Corresponding Point Input
Select the inputted images via the function (1). Enlarge the images
via the function (5). Input the corresponding points at 1:1 scale
accuracy via the function (6). Detect the placement error from the
corresponding points. (D) Manual Placement Error Detection Select
the inputted images via the function (1). Manually detect the
placement error by manually aligning the corrected images via the
function (9).
The present invention has been described in conjunction with a
radiographic imaging device which takes a long image of a subject
in the standing position. However, the present invention is not
limited to use with such devices, and may also be applicable to
radiographic imaging devices which take a long image of a subject
in the supine position.
* * * * *